we calculate annual series of adjusted covered nominal interest rate differentials (see Appendix B):
j = 12
∑r . . * Euro , j∙ t + 3 ∙. η
Year
Domestic
[lr,i + 3-lf,l + 3- (ft ^,s⅛)lj
(9)
J = I
12
which is equivalent to the closed nominal interest rate differential under the assumption of covered interest
rate parity in offshore markets. Henceforth, when we speak about closed interest differentials we also mean
adjusted covered interest rate differentials.
Appendix C plots the year-by-year average deviations from the price measure over the period 1973-1993.
Unfortunately, we had to take a shorter sample period for Italy (1977-1993) due to the lack of forward
exchange rate data. Clearly, Appendix C shows a declining pattern of closed interest differentials, with
alternating periods of relatively high and low capital control intensity. Finally, note that care must be exercised
in taking the closed interest rate differentials as an indication of the intensity of capital controls, since
differences in asset-specific types of risk cannot be completely excluded.11 The next section identifies
the main determinants of capital controls. Why do governments regulate financial markets?
3 The rationales for capital controls
This section focuses on the rationales for capital controls. Several authors have dealt with the issue of the
rationales for capital controls (see e.g. Cairncross, 1973, OECD, 1980, OECD, 1990a, Mathieson and Rojas-
Suarez, 1993, 1994 and the sections on capital flows of IMF’s Annual Report on Exchange Arrangements
and Exchange Restrictions). We take an eclectic approach to identify the main determinants of financial
integration in the EU. Thus, we investigate whether certain economic, institutional and political features
of individual countries may explain the intensity of capital controls.12 In addition, we offer some analytical
indicators for the determinants of capital controls. Some determinants of capital controls may actually apply
more to long-term capital flows. So, in the empirical analysis we should find them to be less significant.
According to Mathieson and Rojas-Suarez (1994) the main three arguments for capital controls in industrial
countries are: (1) Limiting volatile short-term capital flows, (2) Retaining domestic savings and (3)
Maintaining the domestic tax structure (tax rates and tax base). Cairncross (1973) comes up with a fourth
rationale for capital controls: (4) Applying the "second-best" principle of welfare economics.
(1) Limiting volatile short-term capital flows
(1a) Capital controls support the sustainability of fixed but adjustable c.q. unilateral pegged exchange rates:
i.e. capital controls limit the scope for speculative attacks13
The recent turbulence in the European Monetary System (EMS) of pegged but adjustable exchange rates
and of countries unilateral pegging their exchange rates to the ECU orDM (Sweden and Finland), highlights
the importance of the choice of the exchange rate arrangement. An important feature of the functioning
of the EMS was the maintenance of many capital controls to support monetary policy. Controls took a
variety of forms, ranging from taxes on holdings of foreign-currency assets to restrictions on the ability
11 The price measure may be more informative about country-specific regulation using treasury bill rates with the same default
risks. Unfortunately, treasury bill rates are unavailable for all EU countries considered, or the depth of the treasury bill market
is low (see Cumby and Obstfeld, 1984, p. 132 and Appendix B).
12 The rationales for capital controls are often intimately related. So, they may be encompassed under more headings.
13 This subdivision partially follows Alesina, Grilli and Milesi-Ferretti (1994).